kfp-api: test_model_controller_migration times out — migrated model never appears on destination controller (juju:3/stable, ubuntu:24.04)

**Note: This issue was generated with AI assistance (GitHub Copilot) based on automated log analysis and triage.**
Filed by @canonical/solutions-qa

## Summary
`test_model_controller_migration` fails for `kfp-api` with the migrated model never appearing on the destination controller within the 15-minute wait.

## Affected artefact
- Charm: kfp-api
- Revision/track: 2759, track 2.16/edge
- Environments: `juju:3/stable ubuntu:24.04 cmr/kubernetes/kubernetes` and `juju:3/stable ubuntu:24.04 kubernetes`

## Failing test
`test_model_controller_migration`

## Observed error (verbatim io_log, execution 658370)
```
INFO Migrating model 'charmqa-30422712230-40876414' from source controller 'charmqa-30422712230-07861380' to target controller 'charmqa-30422712230-55925593'
INFO Waiting Waiting 0:15:00 for model charmqa-30422712230-55925593:charmqa-30422712230-40876414 to exist before continuing.
ERROR Failure in test_model_controller_migration: juju.backend.JujuWaitTimeoutError: Timed out while Model charmqa-30422712230-55925593:charmqa-30422712230-40876414 does not exist (insufficient status checks)
```

## Scope
- 17 FAILED `test_model_controller_migration` results for `kfp-api` since 2026-07-28, all sharing this exact error signature (only the model/controller names differ).

## Related prior issue
[#734](https://github.com/canonical/charm-integration-testing/issues/734) reported an identical top-level error message (`JujuWaitTimeoutError: ... does not exist (insufficient status checks)`) and was closed as completed on 2026-07-08 after a root-cause analysis pointing to a `juju/description` panic triggered specifically by `mysql-k8s` revision 426+ container `uid`/`gid` metadata on `ubuntu:26.04`. These `kfp-api` failures are on `ubuntu:24.04` and do not involve `mysql-k8s`, so the previously identified trigger condition does not apply here — recording this as a new/possibly-regressed occurrence of the same symptom rather than assuming it's covered by the prior fix.

## Steps to reproduce
1. Trigger the charm QA pipeline for `kfp-api` revision 2759 (track 2.16/edge) on `juju:3/stable`, `ubuntu:24.04`, kubernetes (with or without cmr).
2. Run `test_model_controller_migration`.
3. Observe the migration wait time out after 15 minutes with the model never appearing on the destination controller.

## Test Observer
https://test-observer.canonical.com/#/charms/409557?testExecutionId=658370&testResultId=12630868

Start with test_model_controller_migration and reproduce execution 658370 using the listed Juju 3, Ubuntu 24.04, and Kubernetes environments. Compare the migration wait behavior with issue #734, then trace why the model never appears on the destination controller; done means identifying the regression or failure cause and restoring the test pass.

Written by the indexing model from the issue text.